Methods

An inspection survey of helmet use was conducted in July and August 1998. July and August are usually the hottest months in Greece. Teams consisting of one of three trained interviewers, staff members of the Centre for Research and Prevention of Injuries among the Young (CEREPRI) and one road traffic policeman, randomly stopped TWMVs in 10 sites of randomly selected secondary roads (having one lane for each direction), five systematically selected main road arteries (which run inside the city and suburban area and have two or more lanes for each direction), and five sites on the principal highways linking the capital of Athens with the rest of the country. According to the Hellenic Road Traffic Police Department this scheme allows a reasonable approximation to random sampling, given the different traffic density in the road system of the Greater Athens area. During the survey, 982 TWMVs were stopped and all 1331 involved TWMV riders were interviewed. After the TWMV was stopped, the interviewer ascertained helmet use by the TWMV riders and completed a structured questionnaire with a few open ended questions concerning basic sociodemographic data as well as vehicle related variables. The interview ended with recommendations for helmet use at all times. In no instance were the riders fined for non-use of helmet or any other road traffic violation during this survey.

For the analysis, simple cross tabulations were initially used; subsequently the data were modeled through multiple logistic regression to evaluate factors affecting helmet use, controlling for possible confounding by other variables.^{25, 26}

Results

Table 1 presents the distribution of the 1331 TWMV riders by helmet use and, alternatively, by a series of the studied variables. These data incorporate confounding and should be interpreted with caution. Overall, only 20.2% of TWMV riders were using helmets of any type. The prevalence of use ranged from 9.7% on small suburban roads to 50.8% on highways. Older and female TWMV riders appear to be more frequent helmet users, as are riders of TWMVs with higher engine capacity. There is also some evidence that helmet use is more frequent during day rather than night hours and during weekdays rather than weekends. There were very few unlicensed drivers (4%) but a substantial number of passengers were unlicensed (82%) and they were less likely to use a helmet compared with those who were licensed. A considerable proportion of riders (19%) claimed that use of a helmet increased rather than decreased the risk of an injury by reducing their field of vision, creating discomfort, etc.

Table 2 presents multiple logistic regression derived odds ratios for helmet use associated with a series of potential risk factors. This table allows a better assessment of mutually independent associations between several variables, on the one hand, and helmet use on the other. The odds for helmet use were 10 times higher in highways in comparison with small suburban roads. During the weekends and night hours the prevalence of helmet use was significantly lower. Passengers were less likely than drivers to wear helmets and, after controlling for riding position, the odds of a woman using a helmet were almost twice those of a man. Riders of larger engine capacity TWMVs were significantly more likely to wear a helmet. Helmet non-users were more likely to claim that they were not convinced about the safety benefits conferred by helmets. No statistically significant association was found for helmet use with age and time since obtaining a TWMV driving license, when the other indicated factors were controlled for in the regression model.

Among helmet non-users, the majority (46%) indicated that “the helmet made them feel uncomfortable”, particularly in warm weather. Altogether 18% claimed that there was little need for helmet use in low speed riding. Fewer than 5%, but more than 1% of helmet non-users attributed non-use to such reasons as (in decreasing frequency): “the helmet was too expensive to buy”, “it would be stolen during small errands or entertainment in the city”, they “forgot it”, it disturbed “the vision”, “hearing”, or “the shape of hairstyle”, etc. Fewer than 1% of non-users claimed that they were unable to wear one because they had to make frequent stops on account of their job (letter and package couriers), or that they did not want to look like terrorists, or they invoked strange health reasons. A notably large number (26 riders) claimed that they did not like the helmet or that they actually felt uncomfortable when wearing it, but they nevertheless complied with the law that required carrying rather than wearing a helmet!

Discussion

This study has several advantages. The sample was reasonably large and designed to be representative. There were essentially no refusals and the information elicited during the interviews allowed for a better documentation of factors affecting the use or non-use of helmets in comparison with simple observational surveys. Moreover, the study was the first of its kind undertaken in Greece. It illustrates a major public health problem that underlies the high incidence of injuries among TWMV riders in this country.²⁴ Although none of the results was unexpected, the extremely low prevalence of use, and the quantification of the patterns of non-use should allow rational planning of measures to increase use of these protective devices. It is possible that the sampling method used did not generate representative estimates in the strict statistical sense but all our results were presented conditioned on the possible selection factors (age, gender, type of road, etc) and moreover, the overall results were too striking to be explained by minor biases.

It has been established that helmets provide considerable protection in TWMV crashes.^12–17 In a study in Greece, the risk of death associated with helmet use was reduced by 36%.²⁴ The high relative risk for injuries during TWMV crashes, associated with helmet non-use, generates a high attributable risk, because transportation by TWMVs is so common. Moreover, TWMV use is increasing in Greece, particularly in large cities, such as Athens, but also in the isolated parts of Greek islands and other urban settings. There are currently some 600 000 motorbikes in Athens, mostly of small engine capacity and mostly used by younger people. It is sad that when the Scandinavian countries have succeeded in achieving almost universal use of helmets by TWMV riders, and are now trying to extend the use of helmets to bicyclists as well, the use of helmets among TWMV riders in Greece is limited to fewer than a quarter of TWMV riders, at least during the hot summer months of July and August.

Several years ago Geoffrey Rose pointed out a public health paradox.²⁷ Those relatively few people at high risk for an untoward event—be it stroke among those with high blood pressure, heart attack among those with very high concentrations of blood cholesterol, or head injuries among high speed TWMV riders—are likely to benefit most from the application of respective preventive measures. Therefore, they are highly motivated to adopt the proposed preventive measures. Nevertheless, the attributable risk, which best reflects a public health problem, is mostly contributed to by the far larger number of those who are at moderately increased risk—in this instance, medium speed TWMV riders in city conditions. In our study, more than 50% of those riding in highways were using a helmet, whereas the prevalence of helmet use was less than 15% among the far larger number of TWMV users riding in main city and suburban roads. A substantial fraction of injuries and deaths could be avoided if the latter group was targeted with the same intensity.

The results of this investigation also highlight other important problems. The likelihood of helmet use is lower during weekends, when many crashes occur, and during the night hours, when the severity of these injuries is generally higher. Male drivers and passengers are more frequently risk takers in comparison with their female counterparts, whereas those with a TWMV driving license and/or understanding of the function of a helmet are more likely to use one.^{2, 28, 29} The results and the conclusions derived from studying these factors provide the necessary documentation for the priorities for campaigns or measures targeting extensive helmet use by TWMV riders.^{5, 29} In addition, personal complaints, such as those concerning the field of vision, hearing ability, or hairstyle, should be taken into account.^30–32

Reasons for non-use of helmets are revealing. Helmets have been primarily and extensively used in North American and Northern European countries, so that their design did not take into account the hot weather conditions prevailing in the Mediterranean countries. Moreover, most TWMV riders in Greece ride small engine capacity mopeds, which are relatively inexpensive. For the young and financially stressed TWMV rider, it may appear irrational to pay an extra 20% to purchase a standard helmet in addition to the light TWMV he has just managed to buy, or to even think about replacing the helmet according to the indicated date of expiration.^{33, 34}

Epidemiology and preventive medicine have been criticized for having gradually lost their social origin and public health perspective.^{35, 36} This may or may not be generally true but there are some situations in which it certainly is. Mortality and morbidity from TWMV injuries are major public health problems in many countries, including Greece and other Mediterranean countries.^{24, 37} Thus, it is sad to realize how little attention TWMV injuries have received in the mainstream epidemiologic research. Our results indicate that there is an urgent need for a multipronged campaign in Greece to increase the prevalence of helmet use by TWMV riders. The campaign should include not only police enforcement but also initiatives to make helmets more convenient to wear and less expensive.